From 34d0616e3da89b6b0e0eee27af2f60fd3fa0bb95 Mon Sep 17 00:00:00 2001 From: indaco biazzo Date: Wed, 27 Sep 2023 13:49:08 +0200 Subject: [PATCH] Update README.md --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 7d06c00..5787aaf 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,6 @@ # H2ARNN ## Pairwise interacting systems (H2) AutoRegressive Neural Network +[![DOI](https://zenodo.org/badge/196019537.svg)](https://zenodo.org/badge/latestdoi/196019537) This repository contains a python package to train Autoregressive Neural Networks (arnn) to learn to generate according to the classical Boltzmann distribution of a generic pairwise interacting spins system. The architectures of the autoregressive neural newtorks implemented so far are the MADE architecture, the CW architecture, the SK-krsb architecture. New architectures can be easily added. This repository contains the code used to generate the images and results presented in the paper: